384 replies to this topic

[Sevronis]

Sable, on 17 August 2015 - 03:29 PM, said:

I guess tryhard means something different to you. To me it's the super competitive crowd that only uses meta builds and tries to perfectly time everything because .00002 seconds extra gives them that edge they think they need to win.

Also at times these same tryhards can be egotistical elitist jerks who may also desire to rub in your face how terribad you were when they win. Not all are like that though, but thats where some of the 'tryhard' as a slur can come from.

[Gas Guzzler]

Boyka, on 17 August 2015 - 03:35 PM, said:

Hi Alex, I think the PSR has to be different for every mech you drive rather then pilot based.. I mean if I doing very well with a mech I love to drive and reaching tier 1, then switching to another mech I'm not very good my team will suffer for my lack of ability with that mech.
Just a though.

Well every mech is a lot of data to keep track of. A Class PSR might be more feasible, but less full-proof because a Vindicator doesn't equal a Stormcrow.

[Void Angel]

On the other hand, the "stomps" that so many people blame on Elo are more often a product of how the game trains people to cower and hide, combined with the cumulative effects of fire superiority. I've lost track of the times I've been hung out to dry by my team because they wouldn't move from their favorite coward holes to deal with enemy flanking movements. In those instances - or just when I make a mistake and die in a pyre of shame - I can get crushed without being able to significantly harm the enemy. Then the ratio of firepower is 11-12, making it that much easier for the next 'mech on my team to die without giving as good as he got, and it snowballs from there.

Since players are so often slow to react (or simply refuse to move,) a successful flanking attack by even-ranked players can turn a standard camping snipefest into a onesided blowout very quickly - it's not necessarily an indication that the matchmaker failed. In any case, the proposed system places hard barriers between high and low-to mid-ranked players, and can be expected to curb the number of truly onesided matches significantly.

[Alex Novian]

Um, so who/when did we vote to have all the matches start out of a dropship like CW? Because that drop sequence is crap.

[bad arcade kitty]

Void Angel, on 17 August 2015 - 03:38 PM, said:

On the other hand, the "stomps" that so many people blame on Elo are more often a product of how the game trains people to cower and hide, combined with the cumulative effects of fire superiority. I've lost track of the times I've been hung out to dry by my team because they wouldn't move from their favorite coward holes to deal with enemy flanking movements.

prepare to have it more

because if they move first they will likely die fast and get a bad personal score!
it doesn't matter if you win or lose if your personal score is bad

Edited by bad arcade kitty, 17 August 2015 - 03:43 PM.

[Void Angel]

PS: Since we don't know how they're weighting any of the secret PSR elements, any predictions that the system will be easily exploitable is unfounded doomsaying. =)

[Deathlike]

Wintersdark, on 17 August 2015 - 02:59 PM, said:

Am I the only one who can read this part?

I did too, but let me clarify later....

Wintersdark, on 17 August 2015 - 03:04 PM, said:

and then right after that they discuss how its weighting the scores. The very next paragraph. Specifically stating that a support/scout could be equally rated with a heavy damage/kill dealer.

Have you ever really looked at the match scoring system as a whole?

One of the tricks I used to get the "easy" 30 points in any of those "match score challenges" was by simply placing a UAV in a decent area and collect the points.

I simply had to do some sub-par damage (probably may have had to use an airstrike to go with that), and I was golden. Easy 30 match point farming. I could do this in any mech, and it's the optimal route for Lights.

I've written already about how this would skew Lights in a bad way... considering the lack of role warfare and the notion that "a good Light" in the game ends up being one that has to do a lot of work compared to its bigger brothers.

When you peek with a Light at higher levels of play... you should be prepared to get smashed if spotted. The most crazy of Light pilots have to really pick and choose when to show themselves. It really does become self-evident.

Void Angel, on 17 August 2015 - 03:16 PM, said:

You do realize that the elements of the match score will be weighted by an internal, secret algorithm, right? It says so right in the patch notes - so all this speculation about damage farming is simply groundless doomsaying.

I kinda treat "secret algorithms" like "security through obscurity". Just because you don't know what it's doing, doesn't actually mean it's doing it correctly or properly.

[Igor Kozyrev]

Can anybody tell me solid numbers for very high, high and low match scores? 150/90/30?

[Wintersdark]

Void Angel, on 17 August 2015 - 03:38 PM, said:

On the other hand, the "stomps" that so many people blame on Elo are more often a product of how the game trains people to cower and hide, combined with the cumulative effects of fire superiority. I've lost track of the times I've been hung out to dry by my team because they wouldn't move from their favorite coward holes to deal with enemy flanking movements. In those instances - or just when I make a mistake and die in a pyre of shame - I can get crushed without being able to significantly harm the enemy. Then the ratio of firepower is 11-12, making it that much easier for the next 'mech on my team to die without giving as good as he got, and it snowballs from there.

Since players are so often slow to react (or simply refuse to move,) a successful flanking attack by even-ranked players can turn a standard camping snipefest into a onesided blowout very quickly - it's not necessarily an indication that the matchmaker failed. In any case, the proposed system places hard barriers between high and low-to mid-ranked players, and can be expected to curb the number of truly onesided matches significantly.

This. I'd be willing to bet cold hard cash that at least 75% of "stomps" had absolutely nothing to do with relative player skill and instead are just because of how the game works.

[Void Angel]

Sable, on 17 August 2015 - 03:29 PM, said:

I guess tryhard means something different to you. To me it's the super competitive crowd that only uses meta builds and tries to perfectly time everything because .00002 seconds extra gives them that edge they think they need to win.

That's how the term is often defined when it's explained, but it's not how it's used, and the underlying mentality of its users is nearly always... lacking, in my experience. It's always trotted out when you're telling people to lasthit, or not break CC, or not to all run off just because there's a fight over there because we need this - oh look, 5 rogues and I'm dead. Anyone telling them the right thing to do is an arrogant, elitist "tryhard," because after all it's "their game," and they'll "play it their way."

Take it from someone who's been on the receiving end - use of the term is often despicable, and you don't want to associate yourself with those people.

[Wintersdark]

Igor Kozyrev, on 17 August 2015 - 03:44 PM, said:

Can anybody tell me solid numbers for very high, high and low match scores? 150/90/30?

Over 100 is a pretty damn good game. Basically, once you're exceeding 100 points, you're probably the only one on the team doing it; at most, one more. You don't just see ever higher match scores as you improve - generally, getting really high match scores is either because you're doing a lot of wasted damage (that is, damage that doesn't contribute to a kill) or because your teammates where underperforming or the stars aligned to allow you to encounter armor stripped but largely intact enemies. These are easy kills, and grant a great deal of bonus points in component destruction.

I tend to get ridiculously high scores in my DWF's because of that. I'm slow, but I'm pushing a 120 point alpha. By the time I get stuck in close, everyone is damaged, and I can just tear through mech after mech, building an awesome score and looking like a hero... While realistically, I'd have been able to contribute more if my teammates could have used my armor to soak some damage and such... then they'd have had better scores (having survived longer) at my expense, but the team would have done better.

And that right there is what a lot of people are concerned about with this new system.


But, yeah. I think 30pts (loss) or 50pts (win) is what you should strive for as a "I did my job" score, and look for 60/80 as "I did well." 20 points is the flat "win" bonus.

[Void Angel]

Deathlike, on 17 August 2015 - 03:44 PM, said:

Have you ever really looked at the match scoring system as a whole?

One of the tricks I used to get the "easy" 30 points in any of those "match score challenges" was by simply placing a UAV in a decent area and collect the points.

I simply had to do some sub-par damage (probably may have had to use an airstrike to go with that), and I was golden. Easy 30 match point farming. I could do this in any mech, and it's the optimal route for Lights.

I've written already about how this would skew Lights in a bad way... considering the lack of role warfare and the notion that "a good Light" in the game ends up being one that has to do a lot of work compared to its bigger brothers.

When you peek with a Light at higher levels of play... you should be prepared to get smashed if spotted. The most crazy of Light pilots have to really pick and choose when to show themselves. It really does become self-evident.


I kinda treat "secret algorithms" like "security through obscurity". Just because you don't know what it's doing, doesn't actually mean it's doing it correctly or properly.

This is silly. You do not have the data to extrapolate exploits from the current matchmaking system - thus all of your reasoning is invalid. Misapplying cyber-security philosophies to this subject doesn't strengthen your point, either. In order to exploit the system, an attacker would have to first extrapolate the system's hidden algorithms. This would be extremely hard without a huge data set to analyze - and possibly impossible if they don't make tiers visible. The problems are non-trivial, and any exploiters would be a tiny minority of the player base.

All of this is pure speculation on false premises.

[Richter Kerensky]

I'm concerned about the idea that the system won't expand to include more tiers to be valid to match you with the longer you wait. Did I read this correctly? Does this game have a large enough player base to have a hard limitation like this? If you play during off-peak hours are you just never going to get a match?

Edited by Richter Kerensky, 17 August 2015 - 03:57 PM.

[Void Angel]

Richter Kerensky, on 17 August 2015 - 03:56 PM, said:

I'm concerned about the idea that the system won't expand to include more tiers to be valid to match you with the longer you wait. Did I read this correctly? Does this game have a large enough player base to have a hard limitation like this? If you play during off-peak hours are you just never going to get a match?

It didn't say tier 2 and 3 wouldn't get matched with tier 1 - it said Tiers 4 and 5 wouldn't. So I think they're expecting the population to be able to handle it without a rehash of the match failures on the top (and bottom) of the Elo scale which occurred in the early matchmaker.

[Deathlike]

Void Angel, on 17 August 2015 - 03:56 PM, said:

This is silly. You do not have the data to extrapolate exploits from the current matchmaking system - thus all of your reasoning is invalid. Misapplying cyber-security philosophies to this subject doesn't strengthen your point, either. In order to exploit the system, an attacker would have to first extrapolate the system's hidden algorithms. This would be extremely hard without a huge data set to analyze - and possibly impossible if they don't make tiers visible. The problems are non-trivial, and any exploiters would be a tiny minority of the player base.

All of this is pure speculation on false premises.

I don't think you paid attention to the actual PREMISE of the new system.

Do I have to quote things again?

The focus is on match score - something I'm familiar with and know enough how to manipulate it favorably (or unfavorably I suppose).

Quote

As a replacement for the original Elo-based rating and matching system, the new Pilot Skill Rating system releasing in this patch will provide more accurate and diverse skill ratings to better reflect your individual level of performance match-to-match. Whereas your original Elo scores were determined solely by your win/loss rate in a given weight class, the new PSR system is primarily influenced by your Match Score. In other words, any player action in the game that triggers a C-Bill or XP reward now applies to this new scoring system.

I don't even know how not to repeat what I said earlier.

[Void Angel]

...
You're embarrassing yourself.

InnerSphereNews, on 17 August 2015 - 01:31 PM, said:

We use the action triggers in a weighted formula that employs multipliers to place greater emphasis on certain actions over others.

...

The weighted formula and action triggers will not be released publicly.

You don't have. The data you need. To make the claims. You are making.

Wanna make any more snide remarks about reading comprehension?

Edited by Void Angel, 17 August 2015 - 04:08 PM.

[KharnZor]

This should shake things up a bit..
Then there's the big balance pass coming soon(ish)
Possible interesting times ahead

[Void Angel]

InnerSphereNews, on 17 August 2015 - 01:31 PM, said:

We use the action triggers in a weighted formula that employs multipliers to place greater emphasis on certain actions over others.The weighted formula and action triggers will not be released publicly.

[Wintersdark]

KharnZor, on 17 August 2015 - 04:03 PM, said:

This should shake things up a bit..
Then there's the big balance pass coming soon(ish)
Possible interesting times ahead

This definitely looks like it's designed to tie into the new balance pass.

Note the reference to new scouting and support actions? What could they be? Could they finally be expanding role warfare? Oh, I can dream.

[Deathlike]

Void Angel, on 17 August 2015 - 04:02 PM, said:

...
You're embarrassing yourself.

Why don't you explain it?

The premise is stated as is, so I'm not sure what you are looking for.

I'm not saying this would be abused exactly, but it will causes certain sub-sections of the playbase by design to not function properly with this system.

If that is the premise and you understand the basics of how the match score works, then there will be certain results that are going to happen.

I'm not saying I'm right, but I know enough about the system (not necessarily the other logistics) to know it's going to function a certain way.